See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/359090295

Storing information of stroke rehabilitation patients using blockchain

technology: a software study

Article in Yeungnam University Journal of Medicine · March 2022

DOI: 10.12701/yujm.2021.01368

CITATIONS READS

5 41

1 author:

Min Cheol Chang

College of Medicine, Yeungnam University
507 PUBLICATIONS 5,763 CITATIONS

SEE PROFILE

All content following this page was uploaded by Min Cheol Chang on 08 March 2022.

The user has requested enhancement of the downloaded file.

Original article
eISSN 2799-8010
J Yeungnam Med Sci 2022;39(2):98-107
https://doi.org/10.12701/yujm.2021.01368

Storing information of stroke rehabilitation patients

using blockchain technology: a software study
Min Cheol Chang
Department of Physical Medicine and Rehabilitation, Yeungnam University College of Medicine, Daegu, Korea

Background: Stroke patients usually experience damage to multiple functions and a long rehabilitation period. Hence, there is a large
volume of patient clinical information. It thus takes a long time for clinicians to identify the patient’s information and essential pieces
of information may be overlooked. To solve this, we stored the essential clinical information of stroke patients in a blockchain and im-
plemented the blockchain technology using the Java programming language.
Methods: We created a mini blockchain to store the medical information of patients using the Java programming language.
Results: After generating a unique pair of public/private keys for identity verification, a patient’s identity is verified by applying the El-
liptic Curve Digital Signature Algorithm based on the generated keys. When the identity verification is complete, new medical data are
stored in the transaction list and the generated transaction is verified. When verification is completed normally, the block hash value
is derived using the transaction value and the hash value of the previous block. The hash value of the previous block is then stored in
the generated block to interconnect the blocks.
Conclusion: We demonstrated that blockchain can be used to store and deliver the patient information of stroke patients. It may be
difficult to directly implement the code that we developed in the medical field, but it can serve as a starting point for the creation of a
blockchain system to be used in the field.

Keywords: Blockchain; Information services; Patients; Rehabilitation; Stroke

Introduction without a central server, and the content can be trusted without a
third-party guarantee. The applications of blockchain have been
A blockchain, also called a distributed or shared ledger, is a tech- expanding to various fields including government, finance, and pub-
nology through which participants jointly verify, store, distribute, lic data, and its use is also being explored in the medical field [3-5].
and interconnect data without an authorized third party by gener- Currently, patient medical information is stored on hospital
ating data in blocks [1,2]. It allows participants to jointly record servers, and hospitals cannot easily exchange the patient informa-
data by distributing the data to a person-to-person (P2P) network tion stored on such servers with other hospitals because of the risk
rather than to the central server of a specific organization [1,2]. of a privacy breach [1]. Consequently, when patients are trans-
Those who are permitted to see the ledger read it according to an ferred to another hospital, they need to carry their medical infor-
agreed method, and the transactions are also recorded according to mation from one hospital to the next. Conversely, if blockchain
an agreed method. In this way, information can be stored securely were used in place of servers, the information would be generated

Received: July 28, 2021 • Revised: August 13, 2021 • Accepted: August 13, 2021
Corresponding author: Min Cheol Chang, MD
Department of Physical Medicine and Rehabilitation, Yeungnam University College of Medicine, 170 Hyeonchung-ro, Nam-gu, Daegu 42415, Korea
Tel: +82-53-620-4682 • Fax: +82-53-4231-8694 • E-mail: [email protected]

Copyright © 2022 Yeungnam University College of Medicine, Yeungnam University Institute of Medical Science
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/)
which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

98
 J Yeungnam Med Sci 2022;39(2):98-107

and recorded in block units and then stored on multiple nodes in a or process. However, the format in which the key data are encoded
distributed manner [6]. This would make hacking practically im- and managed can be conveniently viewed and transmitted using
possible while allowing hospitals to easily and freely share patient the Base64 algorithm (an encoding algorithm that converts binary
information with other hospitals. data to text). An encoded file is called ‘Privacy-Enhanced Mail’ and
During the rehabilitation of a stroke patient, a large amount of generally has the file extension ‘.pem’. The keys are randomly gen-
data related to the patient’s condition is generated due to the long erated using the chart number or resident registration number of
rehabilitation period [7]. Furthermore, many patients are trans- the patient.
ferred to different hospitals to receive rehabilitation treatments. Code: Generation of private and public keys
When a patient moves to another hospital, the patient or their A unique pair of private/public keys are generated for a given pa-
guardian must receive the printed patient care information from tient using the Elliptic Curve Digital Signature Algorithm (ECD-
the old hospital and deliver it to the new hospital [1,7]. Doctors SA) algorithm. The generated keys have the file extension ‘.pem’
check the current and past conditions of transferred patients based and are stored under a specific file name.
on these paper records, which takes a considerable amount of time.
Moreover, important pieces of information may be missed due to
Generation of private and public keys using the Elliptic
the large volume of information and because each hospital records
Curve Digital Signature Algorithm
patient information in its own way. We believe that blockchain can
help to solve this problem. // Private and public keys are generated using the ECDSA algo-
Based on the previous study [7], we identified the essential in- rithm and then stored.
formation for stroke patients receiving rehabilitation treatment and // For a detailed version of the ECDSA, we use sect163k1.
then used private blockchain network technology to store the in- public void generate (String privateKeyName, String publicKey-
formation of one patient through the medical information transac- Name) throws Exception {
tion process. Since the personal medical information of patients // The ECDSA algorithm of the Bouncy Castle is used.
can be transferred on the network, only the participants who were KeyPairGenerator generator = KeyPairGenerator.getInstance("-
authorized to use the private blockchain were allowed to write and ECDSA", "BC");
read the patient medical information. // sect163k1 is the algorithm used to generate the elliptic curve.
ECGenParameterSpec ecsp;
Methods ecsp = new ECGenParameterSpec("sect163k1");
generator.initialize(ecsp, new SecureRandom());
We created a mini blockchain to store the medical information of // A random pair of keys is generated using this algorithm.
patients using the Java programming language. To actually run the KeyPair keyPair = generator.generateKeyPair();
source code written in Java, we installed the Java Development Kit System.out.println("A pair of elliptic curve encryption keys was
and Eclipse, the most representative integrated Java development generated.");
environment. // The private and public keys are extracted from the generated
keys.
Results PrivateKey priv = keyPair.getPrivate();
PublicKey pub = keyPair.getPublic();
The medical information transaction process is illustrated in Fig. 1. // The private and public keys are stored under specific file names.
writePemFile(priv, "EC PRIVATE KEY", privateKeyName);
1. Patient consent writePemFile(pub, "EC PUBLIC KEY", publicKeyName);
A patient’s consent is required before sharing their personal medi- }
cal information on the network. The procedure for obtaining con-
sent is conducted through an application. Once the patient’s con- // The generated encryption key is save as a file in the .pem class.
sent has been obtained, a unique pair of public/private keys is is- private void writePemFile(Key key, String description, String file-
sued for the patient. The issued keys are stored as files: the private name) throws FileNotFoundException, IOException{
key is stored on the patient’s smartphone, and the public key is Pem pemFile = new Pem(key, description);
stored by the medical institutions that participate in the network. pemFile.write(filename);
In general, keys are stored in a byte format, which is difficult to read (Continued to the next page)

https://doi.org/10.12701/yujm.2021.01368 99
Chang. Blockchain technology in stroke rehabilitation

0. Patient approval process Medical

1. Generation of a transaction institution A
(Generation of the patient's medical information)
2. Generation of a transaction that includes transaction information

on 3.
a cti Tra
ns n sac
Tra
3. on tio
n
ati 6.
Ve
fic
eri ri fic
6.V a tio
n

3. Transaction

6. Verification
Medical Medical
institution B institution D

4. Verify transaction 5. Agreement among 4. Verify transaction

participants

5. g
 on
Ag
ree
Medical
t am ants
institution C en cip
me em arti
pa nt a A gre p
rti mo
cip ng 
an 5.
ts

4. Verify transaction

Fig. 1. Application of blockchain technology to medical information: patient identities are verified using the public key-based structure,
and the medical records of the verified patients are connected as a chain through the hash value.

System.out.println(String.format("Encryption key %s was ex- MEAwEAYHKoZIzj0CAQYFK4EEAAEDLAAEArqRskOO

ported to the file %s.", description, filename)); r0PXRusLWm+8WbSTBXXCBrtH
} x0EfmTpzV/JP8WI5K0YxkwRF
-----END EC PUBLIC KEY-----

When the private and public keys stored in the .pem files are
opened, they have the following format. 2. Identification
The purpose of the identification process is to verify the identity of
patients using a public key-based structure (Fig. 2). The function
Format of private and public keys
of the public key-based structure is to manage passwords as pairs of
-----BEGIN EC PRIVATE KEY----- private and public keys using the ECDSA algorithm. Data encrypt-
MGwCAQAwEAYHKoZIzj0CAQYFK4EEAAEEVTBTAg ed with a private key can only be decrypted with the public key
EBBBUCoeEILPFbMQIh3CRiHo+S with which it is paired. The private key is encrypted using the pa-
3++ka8egBwYFK4EEAAGhLgMsAAQCupGyQ46vQ9dG6w- tient’s chart number or resident registration number and then de-
tab7xZtJMFdcIGu0fHQR+Z crypted by reading the public key. If the patient’s public key or pri-
OnNX8k/xYjkrRjGTBEU= vate key does not exist or has been manipulated or damaged, iden-
-----END EC PRIVATE KEY----- tity verification fails, and it is impossible to access the patient’s
medical information.
-----BEGIN EC PUBLIC KEY----- Code: Identity verification process
(Continued to) The private and public keys of a given patient are read, and their va-

100 https://doi.org/10.12701/yujm.2021.01368
 J Yeungnam Med Sci 2022;39(2):98-107

Block
connection

Block #0 Block #1 Block #2

Block Hash: 0x00dsf.. Block Hash: 0xwek0u.. Block Hash: 0xkidrs..
Gil Previous Hash: 0 Previous Hash: 0x00dsf.. Previous Hash: 0xwek0u..
Dong
Hong Patient name: Gil Dong Hong Patient name: Gil Dong Hong
Chart number: 123456 Chart number: 123456
Birthdate: 1957. 09. 11 Birthdate: 1957. 09. 11
Read medical Treatment date: 2021. 05. 07 Treatment date: 2021. 05. 14
records Institution: A Institution: B
Private Key Public Key Doctor: Kim Doctor: Choi
Treatment details: Treatment details:
Identification Patient 1 Public Key
Gil Dong Hong (key verification) Patient 2 Public Key
Private Key Hong Gil-Dong Public Key Block #5 Block #4 Block #3
…
Block Hash: 0x0dkrk.. Block Hash: 0xwek0u.. Block Hash: 0aakin..
Patient 9 Public Key Add medical
records Previous Hash: 0xwek0u.. Previous Hash: 0aakin.. Previous Hash: 0xkidrs..
(Add a block) Patient name: Gil Dong Hong Patient name: Gil Dong Hong Patient name: Gil Dong Hong
Chart number: 123456 Chart number: 123456 Chart number: 123456
Birthdate: 1957. 09. 11 Birthdate: 1957. 09. 11 Birthdate: 1957. 09. 11
Treatment date: 2021. 06. 11 Treatment date: 2021. 06. 07 Treatment date: 2021. 06. 01
Institution: A Institution: B Institution: C
Doctor: Kim Doctor: Choi Doctor: Hong
Treatment details: Treatment details: Treatment details:

1. Check the existence of the patient's public key

2. Check for abnormality of the key
- If the key is normal → Approve reading of the medical records
- If the key is abnormal → Do not permit reading of the medical records
The validity of the transaction verified and there is consensus among the participants
If the result is normal, the block is connected to the chain

Fig. 2. Encryption process of the public key-based structure. After the consent to use and share personal medical information is provid-
ed, a unique pair of keys (private key, public key) is issued for the patient. An identity verification process is required to read or update
medical records. The process is performed using the unique key pair assigned to the patient. Once the identity is verified normally, the
patient’s past medical records can be read, and new ones can be written. A new medical record is stored in a block, and if no abnormality
is observed through validation verification, the new medical record is linked to the previous block and stored.

Private key

Encrypt private key

Chart number
+
Encrypted data
Resident
registeration number Decrypt public key

Public key

Fig. 3. Distribution and agreement process of medical information transaction. The verification and agreement process for medical in-
formation transaction is shown. The patient’s identity verification process uses a public key-based structure that uses the Elliptic Curve
Digital Signature Algorithm. The public key-based structure is a method of managing passwords with a pair of private and public keys,
and facilitates the encryption of private keys using chart numbers or resident registration numbers. Data encrypted using the private key
can be decrypted by only using the paired public key.

lidity is verified. are extracted. The file is encrypted (signed) using the private key
A text string file (.pem) is read and the private and public keys and decrypted using the public key.

https://doi.org/10.12701/yujm.2021.01368 101
Chang. Blockchain technology in stroke rehabilitation

ed is expressed as a transaction. Transactions are propagated to par-

Verification of the validity of private and public keys
ticipating medical institutions connected to the network. Each par-
// The private and public keys are read and initialized. ticipating medical institution that receives a transaction updates its
public void setFromFile(String privateKey, String publicKey) transaction information and propagates the transaction to another
throws Exception { medical institution. When a medical institution receives a transac-
// A function to extract the private key from the certificate of a text tion, it needs a means to verify whether the transaction is correct. A
string format: digital or electronic signature is required for validity verification,
this.privateKey = new EC().readPrivateKeyFromPemFile(pri- and the original data (the patient’s medical information) is there-
vateKey); fore sent together with electronically signed data to verify that the
// A function to extract the public key from the certificate of a text transaction is correct and that the data has not been modified. Suc-
string format: cessfully verified transaction information is reflected in the hash
this.publicKey = new EC().readPublicKeyFromPemFile(pub- value of the generated block.
licKey); Code: Validity test
} A transaction object that records medical information is generated,
and the validity of the transaction is verified.
// The validity of the key is verified.
public static boolean isKeyVerify(String id, PrivateKey private-
Verification of the validity of transactions containing pa-
Key, PublicKey publicKey) throws Exception{
tients’ medical information
boolean result;
Signature ecdsa; transaction = new Transaction(key, patientsData.getPatientsDa-
Signature signature; ta());
String text;
byte[] baText; // The transaction information to be propagated is created.
byte[] baSignature; // The transaction information includes the patient’s medical in-
// Encryption (signature) using the private key: formation, key information, electronically signed data, and trans-
ecdsa = Signature.getInstance("SHA1withECDSA"); action generation time.
ecdsa.initSign(privateKey); public Transaction(Key key, String patientsData) throws Excep-
text = id; tion {
baText = text.getBytes("UTF-8"); this.patientsData = patientsData
// The data from the original source that are encrypted and signed this.sender = key.getPublicKey();
are output. this.timestamp = Util.getDate();
ecdsa.update(baText); this.signature = key.sign(getData()); // Electronic signature of
baSignature = ecdsa.sign(); original data
// Decryption using the public key when verifying the data: }
signature = Signature.getInstance("SHA1withECDSA");
signature.initVerify(publicKey); // The simple transaction information, excluding the signature
signature.update(baText); value, is returned.
result = signature.verify(baSignature); public String getData() {
return result; return new Util().getHash(sender.toString()) + timestamp + this.
} patientsData
}
3. Reading and adding medical information
After successful identity verification, the medical institution can // The normality of the transaction is verified (a validity test).
read the patient’s past medical records and can update the record public boolean verifyTransaction(Transaction transaction)
with new medical information (Fig. 3). The medical information is throws Exception {
updated through the propagation of a transaction. Moreover, a Signature signature;
medical record can be created, and the message by which it is creat- (Continued to the next page)

102 https://doi.org/10.12701/yujm.2021.01368
 J Yeungnam Med Sci 2022;39(2):98-107

newly generated block that follows it.

signature = Signature.getInstance(“SHA1withECDSA”);
byte[] baText = transaction.getData().getBytes(“UTF-8”);
Generation of blockchain lists using hash values that re-
signature.initVerify(transaction.getSender());
flect transaction information
signature.update(baText);
return signature.verify(new BigInteger(transaction.getSigna-
block = new Block(block.getBlockID()+1, block.getBlock-
ture(), 16).toByteArray());
Hash(), Util.getDate(), patientsData, new ArrayList < Transac-
}
tion> ());

The process for adding medical information to a patient’s medi- // A transaction is generated to add to the medical record of pa-
cal record is as follows: (1) The original data (medical informa- tient 2673123.
tion) is encrypted with the patient’s own private key and electroni- transaction = new Transaction(key, patientsData.getPatientsDa-
cally signed. (2) The original data and the electronically signed ta());
data are propagated to the participating medical institutions. (3) block.addTransaction(transaction);
To verify whether the transaction is valid, the participating medical
institutions decrypt it using the patient’s public key. (4) The de- // The transaction information is reflected in the hash value of the
crypted data and original data are compared to verify the integrity block.// If the transaction information in a block is changed, the
of the data and whether there is any manipulated data. (5) If the re- hash values of all subsequent blocks are also changed.
ceived transaction is determined to be valid, the transaction is up- public String getBlockHash() {
dated in the blockchain, and the transaction is propagated to the // The block hash is created using the previous block hash.
medical institutions participating in the blockchain network. return Util.getHash(getTransaction() + previousBlockHash);
}
4. Block connection // The SHA-256 hash value, which is returned as a text string,
Whenever a transaction of medical information is performed, a passes through the function.
block that contains the transaction information is generated and // When the value of the text string changes, the hash value also
connected continuously to other blocks, and the information is changes.
stored in a distributed manner at the medical institutions partici- // For the SHA-256 hash algorithm, the Avalanche Effect method
pating in the network. Many blocks are closely interconnected is applied.
through the hash values. A hash value is data that is converted to a public static String getHash(String input) {
special text string of a fixed length in which the original data can- StringBuffer result = new StringBuffer();
not be distinguished when the hash goes through the hash func- try {
tion. The transaction information is reflected in the hash value of MessageDigest md = MessageDigest.getInstance("SHA-256");
the newly generated block. When the internal data of a specific md.update(input.getBytes());
block in the blockchain changes, the hash value automatically byte bytes[] = md.digest();
changes, which also affects other blocks. In this way, a blockchain for(int i= 0;i< bytes.lengthi++) {
allows data tampering to be easily detected. In the blockchain, the result.append(Integer.toString((bytes[i] & 0xff) + 0x100, 16).
hash value is used to add the corresponding block to the chain, and substring(1));
the hash value of the previous block is recorded in the current }
block. As a result, a connected list is created in the form of chain. } catch(Exception e) {
Therefore, in order to hack a specific block, it is necessary to tam- e.printStackTrace();
per with other blocks connected to the block of interest, making }
forgery exceedingly difficult. return result.toString();
Code: Generation of block objects }
To create a block object, a verified transaction is added to the
chain, and the transaction information is reflected in the hash value SHA, secure hash algorithm.
of the block. A new block hash is then generated using the previous
block hash. The hash value of the previous block is saved in the

https://doi.org/10.12701/yujm.2021.01368 103
Chang. Blockchain technology in stroke rehabilitation

The patient’s medical information is stored in the generated

Finger extensor: 1
block, and multiple blocks form a list that is connected through
Hip flexor: 2
their hash values.
Knee extensor: 2
Ankle D/F: 1
Blockchain list information
FAC: 1
Aphasia type: conduction aphasia
============================
AQ: 55
Block number: 0
LQ: 48
Created time: 2021-06-14 01:51:36.640
Light tough: .20/12
Previous hash: null
Kinesthetic: .20/12
Block hash: 74234e98afe7498fb5daf1f36ac2d78acc339464f-
MBC: 2
950703b8c019892f982b90b
GCS: 15
Medical record:
============================
============================
A normal transaction was found.
A normal transaction was found.
============================
============================
Block number: 2
Block number: 1
Created time: 2021-06-14 01:51:42.704
Created time: 2021-06-14 01:51:39.661
Previous hash: 6345e8d4fe0b749dfe1862cbfd90817f38d-
Previous hash: 74234e98afe7498fb5daf1f36ac2d78acc339464f-
d3692cd2cd918be58e5becd540944
950703b8c019892f982b90b
Block hash: dbf72d6bf7143b6272308fb0708291c10733c
Block hash: 6345e8d4fe0b749dfe1862cbfd90817f38dd3692cd-
0787553355753fb90e477113a7e
2cd918be58e5becd540944
Medical record:
Medical record:
NAME: Gil Dong Hong
NAME: Gil Dong Hong
AGE: 50
AGE: 50
GENDER: M
GENDER: M
STROKE: Lt. hemiplegia d/t Rt. CR infarct
STROKE: Lt. hemiplegia d/t Rt. CR infarct
ONSET: Jan/01/2021
ONSET: Jan/01/2021
MEP: Jan/12/2021: Rt. ABP 20.6/5600, Lt. ABP NR; Rt. TA
MEP: Jan/12/2021: Rt. ABP 20.6/5600, Lt. ABP NR; Rt. TA
19.5/2200, Lt. TA NR
19.5/2200, Lt. TA NR
SEP: Jan/12/2021 Rt. Median 19.0/24.2, Lt. Median NR; Rt. PT
SEP: Jan/12/2021 Rt. Median 19.0/24.2, Lt. Median NR; Rt. PT
38.2/45.0, Lt. PT NR
38.2/45.0, Lt. PT NR
Evaluation data: Feb/10/2021
Evaluation data: Jan/10/2021
Bathel index: 52
Bathel index: 30
MMSE: 27
MMSE: 24
GDS: 2
GDS: 2
MVPT: 36
MVPT: 30
MFT: .30/10
MFT: .30/8
Purdue: 16/NT
Purdue: 16/NT
Grip Power: .50/8
Grip Power: .50/6
Monofilament: 3.22/5.18
Monofilament: 3.22/5.22
Two point discrimination: .4/6
Two point discrimination: .4/6
Shoulder abductor: 2
Shoulder abductor: 1
Elbow flexor: 3
Elbow flexor: 2
Finger flexor: 3
Finger flexor: 2
(Continued to the next page)
(Continued to)

104 https://doi.org/10.12701/yujm.2021.01368
 J Yeungnam Med Sci 2022;39(2):98-107

Finger extensor: 1 Finger extensor: 2

Hip flexor: 3 Hip flexor: 3
Knee extensor: 3 Knee extensor: 3
Ankle D/F: 1 Ankle D/F: 2
FAC: 2
Aphasia type: conduction aphasia FAC: 3
AQ: 58 Aphasia type: conduction aphasia
LQ: 52 AQ: 58
Light tough: .20/14 LQ: 54
Kinesthetic: .20/14 Light tough: .20/16
MBC: 3 Kinesthetic: .20/16
GCS: 15 MBC: 4
============================ GCS: 15
============================
A normal transaction was found.
============================ A normal transaction was found.
Block number: 3 ============================
Created time: 2021-06-14 01:51:45.737 Block number: 4
Previous hash: dbf72d6bf7143b6272308fb0708291c10733 Created time: 2021-06-14 01:51:48.754
c0787553355753fb90e477113a7e Previous hash: f55e598438047feeafac137016eda953396244cff-
Block hash: f55e598438047feeafac137016eda953396244cff- fe148730499727203e72733
fe148730499727203e72733 Block hash: 5fdfbc0a7d30f454dfc24975805e353ca5912b4d-
Medical record: 81c9875447e6a99644650a8f
NAME: Gil Dong Hong Medical record:
AGE: 50 NAME: Gil Dong Hong
GENDER: M AGE: 50
STROKE: Lt. hemiplegia d/t Rt. CR infarct GENDER: M
ONSET: Jan/01/2021 STROKE: Lt. hemiplegia d/t Rt. CR infarct
MEP: Jan/12/2021: Rt. ABP 20.6/5600, Lt. ABP NR; Rt. TA ONSET: Jan/01/2021
19.5/2200, Lt. TA NR MEP: Jan/12/2021: Rt. ABP 20.6/5600, Lt. ABP NR; Rt. TA
SEP: Jan/12/2021 Rt. Median 19.0/24.2, Lt. Median NR; Rt. PT 19.5/2200, Lt. TA NR
38.2/45.0, Lt. PT NR SEP: Jan/12/2021 Rt. Median 19.0/24.2, Lt. Median NR; Rt. PT
Evaluation data: Mar/13/2021 38.2/45.0, Lt. PT NR
Bathel index: 58 Evaluation data: May/01/2021
MMSE: 28 Bathel index: 60
GDS: 2 MMSE: 28
MVPT: 38 GDS: 2
MFT: .30/12 MVPT: 40
Purdue: 16/NT MFT: .30/12
Grip Power: .50/8 Purdue: 16/NT
Monofilament: 3.22/5.14 Grip Power: .50/8
Two point discrimination: .4/6 Monofilament: 3.22/5.14
Shoulder abductor: 3 Two point discrimination: .4/6
Elbow flexor: 4 Shoulder abductor: 3
Finger flexor: 4 Elbow flexor: 4
(Continued to) (Continued to the next page)

https://doi.org/10.12701/yujm.2021.01368 105
Chang. Blockchain technology in stroke rehabilitation

tion for treatment. We, therefore, added this essential information

Finger flexor: 4
of stroke patients to the blockchain, and the information was
Finger extensor: 2
stored.
Hip flexor: 3
This is the first study to demonstrate the possibility of using
Knee extensor: 3
blockchain for the storage and delivery of the patient information
Ankle D/F: 2
of stroke patients by storing the information in a blockchain. The
FAC: 4
code that we wrote may not be suitable for direct implementation
Aphasia type: conduction aphasia
in the medical field, but it can serve as a foundation for researchers
AQ: 58
to create a blockchain system that can be actually used in the field
LQ: 56
based on future research results.
Light tough: .20/16
Kinesthetic: .20/16
MBC: 4 Notes
GCS: 15
Conflicts of interest
============================
No potential conflict of interest relevant to this article was report-
MEP, motor evoked potential; ABP, abductor pollicis brevis; Rt, right; ed.
Lt, left; TA, tibialis anterior; PT, posterior tibial; NR, no response; SEP,
sensory evoked potential; MMSE, mini-mental state examination; GDS,
global deterioration scale; MVPT, motor-free visual perception test; MFT, Funding
manual function test; NT, not testable; D/F, dorsiflexor; FAC, functional The present study was supported by a National Research Founda-
ambulatory category; AQ, aphasia quotient; LQ, language quotient; MBC,
modified Brunnstrom Classification; GCS, Glasgow Coma Scale. tion of Korea grant funded by the Korean government (grant no.
NRF-2019M3E5D1A02068106).
Discussion
Author contributions
In this study, we created a simple blockchain for the purpose of Conceptualization, formal analysis, investigation, supervision,
allowing hospitals to exchange patient information on a network Writing-original draft, wiring-review & editing: CMC.
in a way that makes hacking practically impossible. The proposed
information sharing method only allows the medical staff of a ORCID
hospital to see a patient’s information stored on the network in Min Cheol Chang, https://orcid.org/0000-0002-7629-7213
the form of a blockchain if there is a private key stored on the pa-
tient’s smartphone. This method circumvents the inconvenience References
of the current system in which patients must print out their med-
ical records and deliver them to the new hospital when being 1. Chang MC, Hau YS, Park JC, Lee JM. The application of block-
transferred. Clinicians may be concerned that due to its decen- chain technology in stroke rehabilitation. Am J Phys Med Reha-
tralized nature, if blockchain is used in the medical field, patients bil 2019;98:e74.
will become the guardians of their medical information. Never- 2. Kuo TT, Kim HE, Ohno-Machado L. Blockchain distributed
theless, in the proposed system, the patients do not directly pos- ledger technologies for biomedical and health care applications.
sess their own medical information. Hence, we expect that the J Am Med Inform Assoc 2017;24:1211–20.
aversion of doctors to information sharing via blockchain will be 3. Chang MC, Hsiao MY, Boudier-Revéret M. Blockchain tech-
small. nology: efficiently managing medical information in the pain
This study only included the information that is required by cli- management field. Pain Med 2020;21:1512–3.
nicians to treat stroke rehabilitation patients in the blockchain. The 4. Dutta P, Choi TM, Somani S, Butala R. Blockchain technology
information was based on the paper by Kim et al. [7] published in in supply chain operations: applications, challenges and re-
2021. Through a Delphi study, they collected the essential medical search opportunities. Transp Res E Logist Transp Rev 2020;
information of stroke rehabilitation patients when transferred to 142:102067.
another hospital from 31 physiatrists. They found that the degree 5. Karandikar N, Chakravorty A, Rong C. Blockchain based trans-
of muscle strength at major joints; a brief cognitive function test; action system with fungible and non-fungible tokens for a com-
and hand, language, and sensory function were essential informa- munity-based energy infrastructure. Sensors (Basel) 2021;

106 https://doi.org/10.12701/yujm.2021.01368
 J Yeungnam Med Sci 2022;39(2):98-107

21:3822. 7. Kim JK, Hau YS, Kwak S, Chang MC. Essential medical infor-
6. Fang HS, Tan TH, Tan YF, Tan CJM. Blockchain personal mation for stroke patients undergoing interhospital transfer: a
health records: systematic review. J Med Internet Res 2021; Delphi study. Am J Phys Med Rehabil 2021;100:354–8.
23:e25094.

https://doi.org/10.12701/yujm.2021.01368 107

View publication stats